Helena Britze

EpiRanha identifies epitope similarities using a structure- and residue-aware graph neural network, improving detection of cross-reactive epitopes and off-target interactions.

The paper disentangles protein-interaction and protein-quality effects in antibody-antigen binding using deep mutational scanning and evaluates models for their ability to predict these effects.

The paper creates a dataset of antibody-antigen complexes with mutations and evaluates affinity prediction methods, finding that current models poorly predict binding effects compared to empirical methods.